CN115865220B - AUV auxiliary underwater sound network data transmission method based on intelligent super surface - Google Patents
AUV auxiliary underwater sound network data transmission method based on intelligent super surface Download PDFInfo
- Publication number
- CN115865220B CN115865220B CN202211471729.8A CN202211471729A CN115865220B CN 115865220 B CN115865220 B CN 115865220B CN 202211471729 A CN202211471729 A CN 202211471729A CN 115865220 B CN115865220 B CN 115865220B
- Authority
- CN
- China
- Prior art keywords
- information
- ris
- auv
- sub
- importance
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 230000005540 biological transmission Effects 0.000 title claims abstract description 26
- 238000000034 method Methods 0.000 title claims abstract description 24
- 238000004364 calculation method Methods 0.000 claims abstract description 17
- 238000004891 communication Methods 0.000 claims abstract description 16
- 238000005457 optimization Methods 0.000 claims abstract description 14
- 238000002360 preparation method Methods 0.000 claims abstract description 6
- 230000033228 biological regulation Effects 0.000 claims abstract description 5
- 239000011159 matrix material Substances 0.000 claims description 46
- 230000006870 function Effects 0.000 claims description 19
- 238000004422 calculation algorithm Methods 0.000 claims description 16
- 230000000903 blocking effect Effects 0.000 claims description 6
- 239000003795 chemical substances by application Substances 0.000 claims description 6
- 238000012549 training Methods 0.000 claims description 6
- 238000013461 design Methods 0.000 claims description 5
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 4
- 238000000638 solvent extraction Methods 0.000 claims description 3
- 230000007613 environmental effect Effects 0.000 claims description 2
- 230000007774 longterm Effects 0.000 claims description 2
- 238000010586 diagram Methods 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 5
- 238000005265 energy consumption Methods 0.000 description 4
- 238000011161 development Methods 0.000 description 3
- 230000018109 developmental process Effects 0.000 description 3
- 230000009191 jumping Effects 0.000 description 3
- 238000009825 accumulation Methods 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 230000006399 behavior Effects 0.000 description 1
- 238000005094 computer simulation Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000002787 reinforcement Effects 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Landscapes
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
An AUV auxiliary underwater sound network data transmission method based on intelligent super surface relates to underwater communication transmission. Determining the central position of an energy cavity and the information importance of information to be transmitted in an underwater acoustic communication network, and the position of a target node; the AUV carries an intelligent super surface and forwards the AUV to an energy cavity center for relay; the information source node sends the prepared data sub-packet to the AUV, and the AUV performs RIS system block optimization according to the data sub-packet information, so as to realize the balance between the calculation complexity based on the information importance and the beam forming capability; the RIS completes phase regulation according to a block optimization scheme, and the AUV transmits a preparation signal to the information source node; the information source node integrates and sends the information sub-packets to be transmitted to the AUV, the information sub-packets are reflected to a plurality of destination nodes through RIS, and the destination nodes process the signal segments belonging to the destination nodes according to the signal labels. And through the block forwarding of the RIS system, the simultaneous forwarding of a plurality of information in the underwater acoustic communication network energy cavity is realized.
Description
Technical Field
The invention relates to underwater communication, in particular to an AUV auxiliary underwater sound network data transmission method based on an intelligent super surface.
Background
With the development and utilization of the ocean, the underwater communication technology is an indispensable key technology for the development of the ocean industry such as the ocean oil development, the ocean fishing and transportation, the ocean bottom military base and the like. The underwater acoustic communication technology is the most reliable underwater long-distance communication mode at present, and is gradually expanded from the application of the initial military field to the civil field, so that new opportunities and challenges are continuously met. The underwater acoustic network nodes are difficult to charge, the node energy is limited, and the nodes are separated from the nodes nearby the Sink node Sink, so that the nodes can consume energy more quickly due to the need of bearing more communication loads, and therefore an unbalanced energy consumption phenomenon is formed, and energy holes appear. In order to solve the problem of energy holes, zhu et al propose an opportunistic routing protocol (R.Zhu,et al,"AReinforcement-Learning-Based Opportunistic Routing Protocol for Energy-Efficient and Void-Avoided UASNs,"IEEE Sensors Journal,vol.22,no.13,pp.13589-13601,2022) based on reinforcement learning to reduce network energy consumption, khan et al propose using an underwater robot (AUV) in an underwater acoustic network to collect cruising data, so that network energy consumption is uniformly distributed, the occurrence of the energy holes is avoided .(Khan,et al,"ADistributed Data-Gathering Protocol Using AUV in Underwater Sensor Networks,"Sensors 15,no.8:19331-19350,2015) at present, various optimization algorithms focus on how to avoid the occurrence of the energy holes, and for networks in which the energy holes already occur, but the whole network has more residual energy and needs to be continuously used, the problem of how to continuously transmit data is solved, and the discussion of related documents is less.
Smart supersurface (RIS) is considered one of the key candidate technologies for 6G networks due to its ability to handle the characteristics of the channel environment. RIS has excellent technical characteristics of low power consumption, low cost, easy deployment and the like, and RIS based on electromagnetic wave communication is widely studied in land scenes in recent years (Zhao Yajun, application and challenge [ J ] radio communication technology, 2021,47 (6): 679-691). The RIS concept is equally applicable under water, sun et al propose an acoustic RIS system to significantly increase the data rate of underwater acoustic communications while also maintaining the remote benefits of underwater acoustic communications (Z.Sun,et al.,"High-data-rate Long-range Underwater Communications via Acoustic Reconfigurable Intelligent Surfaces,"IEEE Communications Magazine,vol.60,no.10,pp.96-102,2022).
Under the scene that the network energy consumption is uneven and the energy hole appears, a great deal of manpower resources are consumed for replacing dead sensor nodes, the AUV can be adopted for data relay transmission at the moment, but the time delay of the AUV cruising data collection is longer compared with the time delay of the node acoustic data multi-hop transmission, and the timeliness requirement of emergency information transmission cannot be met. Currently, an AUV is provided with an underwater acoustic RIS system to solve the data transmission problem under the condition that an underwater acoustic network has an energy cavity, and related documents are few and need to be studied. Aiming at the scene that data accumulation cannot be forwarded due to the existence of energy holes in the underwater acoustic network, the invention provides a method for carrying out relay of an RIS system through an AUV, and an information source node can complete information receiving of multiple nodes only by transmitting data once. Because each RIS unit needs to be subjected to beamforming calculation in the beamforming design of the RIS system, the calculation complexity of the reflection waveform of the whole RIS system increases in a cubic manner along with the increase of the number of the RIS units, the RIS needs to be subjected to blocking processing aiming at different information importance degrees, better beamforming capacity is given to data with higher importance degrees, the calculation complexity is reduced, and the AUV relay transmission time is reduced.
Disclosure of Invention
The invention aims to solve the problem of how to continue setting data transmission schemes with different information importance by a node storing data under the condition that an energy cavity exists in an underwater acoustic network, and provides an AUV auxiliary underwater acoustic network data transmission method based on an intelligent super surface. And carrying out data relay work of energy holes by using an AUV (autonomous Underwater vehicle) carried RIS (information processing system), when the information source node stores data with different importance degrees and needs to send the data to different nodes respectively, the RIS carries out RIS block design and beam forming matrix aiming at the data with different information importance degrees and the positions of target nodes, creates specific RIS acoustic reflection paths for different users, realizes the process of simultaneously carrying out underwater acoustic data transmission on different users, and reduces the total RIS beam forming calculation complexity.
The invention comprises the following steps:
1) Determining an area H where an energy cavity is located in an underwater sound network and a source node SN needing to transmit data, classifying the importance IL i of the underwater sound data, wherein the higher the level is, the higher the importance of the representative information is; setting X pieces of information to be transmitted on the SN, respectively transmitting the X pieces of information to be transmitted to X destination nodes, and knowing the importance level of the information to be transmitted; the information source node combines the importance level of each piece of information to be transmitted and the position information of the destination node into a preparation data sub-packet PS;
2) The AUV receives an instruction to run to the center of an area H where the energy cavity is located, and the information source node sends a preliminary data sub-packet PS to the AUV;
3) An NxN RIS unit system is arranged on the AUV, the total area S=NxN of the RIS unit system, and the AUV controls the RIS unit system to carry out block division and RIS reflection path design according to the prepared data sub-packet PS; according to the information importance of X pieces of information to be transmitted, each piece of information corresponds to a sub RIS block S i(i＝1,2,…,X),Si＝Nxi×Nyi (i=1, 2, …, X), the higher the information importance is, the larger the sub RIS block area S i is, and the sum of the areas of all sub RIS blocks is the total area of the RIS system, namely
4) The Q-learning algorithm is utilized to carry out block optimization on X sub RIS blocks, so that the balance between the calculation complexity based on the information importance and the beam forming capability is realized; first, the function Q (s, a) is initialized, and the function is rewardedLearning rate alpha, attenuation coefficient gamma and training frequency K; s represents the blocking state of the whole RIS system; a represents actions selected by an Agent, namely specific block actions for the RIS system; q (s, a) functions represent state action values, namely benefits obtained by the Agent selecting specific block actions in the current state, so that the action selection of each step can be guided; reward function/>The method comprises the steps of representing the benefits obtained by the environmental feedback Agent under different states and actions, namely the benefits of different specific block actions of the RIS system; the learning rate alpha determines the updating speed of the Q matrix; the attenuation coefficient gamma determines the importance degree of decision-making on the gain of long-term decision-making;
4.1 Bonus function) The settings were as follows:
Wherein w 1 represents the bonus weight obtained when the sub RIS block is distributed to the size of S i, w 1 is related to the importance of the information to be transmitted, the higher the importance level of the information is, the stronger the required beamforming capability is, the higher the area of the distributed sub RIS block is, and the larger the bonus weight is; w 1 is calculated as:
w1＝β·ILi
Beta represents the bonus weight based on the importance of the information, w 2 represents the bonus weight subtracted from the sub RIS block due to computational complexity; o (N) represents beamforming computational complexity, related to the number of RIS units allocated;
4.2 Training, randomly selecting an initial state s, randomly generating a white noise matrix WN, adding the white noise matrix WN with a current Q matrix to obtain a temporary selection matrix, selecting an action a 0 corresponding to the maximum value of the temporary selection matrix in all possible actions of the current state s 0, and updating the Q matrix by using the following formula after the action a 0 is selected, wherein the state s 1 is reached:
Q(s,a)←Q(s,a)+α[r+γmaxa′Q(s′,a′)-Q(s,a)]
4.3 Repeating the step 4.2) until training is finished, and obtaining an optimized sub RIS block;
5) The AUV calculates an analog beam forming matrix according to the optimized sub RIS block, and the calculation is completed through the phase shifter, so that the purpose that the sound wave simultaneously transmits information to a plurality of destination nodes through the RIS system is achieved;
6) After the RIS system completes phase regulation, the AUV transmits a preparation signal to an information source node SN, and the SN integrates the information sub-packets to be transmitted and then sends the integrated information sub-packets to the AUV; the information sub-packet is reflected to a plurality of destination nodes through the RIS system, and the destination nodes process the signal segments belonging to the destination nodes according to the signal labels.
The invention considers the difficult problems of energy holes and the accumulation of a plurality of pieces of information to be transmitted with different importance in the underwater acoustic network, proposes the scheme of realizing the directional transmission of a plurality of pieces of data to a plurality of destination nodes by using an AUV to carry RIS, and carrying out the block optimization design of an RIS system by using Q-learning aiming at the information with different importance, thereby realizing the purposes of reducing the calculation complexity, saving the AUV data transmission time and realizing the directional transmission to a plurality of destination nodes.
Compared with the prior art, the invention has the following outstanding advantages:
1) Providing an AUV with an RIS system to solve the problem of energy cavity transmission of the underwater acoustic network, and reflecting signals to a plurality of destination nodes simultaneously by utilizing RIS characteristics so as to save transmission time;
2) Providing RIS system blocking aiming at the importance of underwater sound data so as to reduce the complexity of beamforming calculation;
3) The method comprises the steps of optimizing the blocking problem of the RIS system by using a Q-learning algorithm, and balancing the beamforming capacity and the computational complexity of the RIS system based on the importance degree of information so as to solve the problems that the computational complexity in the beamforming optimization algorithm increases in a cubic manner along with the increase of the number of RIS units and the beamforming capacity of the RIS system increases along with the increase of the number of RIS units.
Drawings
Fig. 1 is a schematic diagram of an AUV assisted underwater acoustic network data transmission method based on an intelligent subsurface.
Fig. 2 is a schematic diagram of underwater node arrangement of an underwater acoustic communication network in the AUV-assisted underwater acoustic network data transmission method based on an intelligent super surface.
Fig. 3 is a schematic diagram of a reward function of different information importance level information under different block actions in the AUV auxiliary underwater sound network data transmission method based on the intelligent super surface.
Fig. 4 is a schematic diagram of Q-value distribution completed by Q-learning segmentation in the AUV-assisted underwater acoustic network data transmission method based on intelligent super-surface according to the present invention.
Detailed Description
The invention is described in detail below with reference to the drawings and the specific embodiments.
As shown in the scenario of fig. 1, an embodiment of the present invention includes the following steps:
1) The embodiment sets a 1000×1000 two-dimensional underwater acoustic network scenario, including s=50 sensor nodes and a source node, where the source node SN is located at coordinates (500,0). The range H of the energy cavity in the underwater acoustic network scene is determined, and in this embodiment, H is set to be a circular range with a radius of 300m by taking (500 ) as a center. The information importance degree IL i is divided into five levels, that is, IL i =1, 2,3,4,5, and the higher the level is, the more important the representative information is.
2) The AUV was driven to the center of the energy cavity H, and a 60X 60 RIS system was provided for the AUV. The SN has x=3 pieces of information to be transmitted, and the information importance of the three pieces of information X 1,X2,X3 is IL i＝1,ILi＝3,ILi =5, which are to be transmitted to destination nodes (250, 1000), (500, 1000), (750, 1000) respectively. The SN combines the importance of the information to be transmitted and the location information of each destination node into a preliminary data sub-packet PS, and transmits the preliminary data sub-packet PS to the AUV.
3) The AUV performs RIS system block optimization through a Q-learning algorithm according to the prepared data sub-packet information, and realizes the balance between the calculation complexity based on the information importance and the beamforming capability. In the Q-learning algorithm, a Q matrix applied to RIS block optimization is an x×n matrix, a row m (m=1, 2, …, X) of the Q matrix represents a serial number of information to be transmitted, a column N (n=1, 2, …, N) of the Q matrix represents the number of RIS units, Q (m, N) represents a Q value corresponding to an action of transmitting data of an mth information to be transmitted through an n×60 sub RIS board, and a larger Q value represents a larger gain of transmitting data of the mth information through the n×60 sub RIS board. Initializing iteration number i=0, wherein the maximum exploration number is K= 7000000, the initial Q value table is a zero matrix of 3×60, and the Q matrix is continuously updated in iteration to record the result after each exploration step and serve as the basis of the final partitioning scheme.
Setting an X N bonus matrixRow m of the bonus matrix (m=1, 2, …, X) represents the information sequence number to be transmitted, column N of the bonus matrix (n=1, 2, …, N) represents the number of RIS units/>Representing rewards obtained by the action of data transmission of the mth piece of information to be transmitted through the n multiplied by 60 sub RIS board; the setting of the bonus matrix is defined by the bonus function/>The decision, representing the trade-off between importance of the information to be transmitted and the complexity of beamforming calculation, is an important basis for updating the Q matrix. The bonus function is calculated as follows:
Wherein w 1 represents the rewarding weight obtained when the sub RIS block is distributed to the size of S i, w 1 is related to the importance of the information to be transmitted, the higher the importance level of the information is, the stronger the beamforming capability is required, the higher the area of the sub RIS block is distributed, and the larger the rewarding is; β represents a bonus weight of importance of information, and β=10 -15 is set in this embodiment. O (N) represents the beamforming computational complexity, and w 2 represents the prize weight subtracted from the sub RIS block due to the computational complexity, in relation to the number of allocated RIS units, w 2 =0.1 is set in this embodiment. In the beam forming optimization algorithm, as the optimization problem is a non-convex problem, the problem needs to be converted into a plurality of sub-problems which are easier to process, a semi-positive relaxation algorithm (SDR) and a sine and cosine optimization algorithm (SCA) are adopted in the sub-problem solving, an alternate optimization algorithm is applied outside the sub-problem, the computation complexity of the SDR algorithm increases along with the number N 6 of RIS units, and the computation complexity of the SCA algorithm increases along with the number N 3 of the RIS units. The computational complexity of the SDR algorithm is taken in this embodiment, i.e., O (N) =n 6.
4) The Q matrix sequentially performs exploration selection of the sub RIS blocks according to the information sequence number to be transmitted, updates the Q value table according to the Q-learning formula, and in this embodiment, takes the learning rate α=0.8, and the attenuation coefficient γ=0.9.
The method comprises the following specific steps:
4.1 Randomly generated white noise matrix WN (1, 60), multiplied by Then adding the first row of the Q matrix, wherein the addition of the Q matrix and the white noise matrix is only a selection basis, the value of the Q matrix is not influenced, and the column number of the maximum value after the addition is selected as the block number s 1 of the information X 1 to be transmitted;
4.2 If s 1 is more than or equal to 58), jumping to the step 4.5); if the current block number s 1 is less than 58, updating the Q matrix according to the objective function, randomly generating a white noise matrix WN (1, 60), multiplying Adding the number of columns of the maximum value after adding to the second row of the Q matrix to be used as the block number s 2 of the information X 2 to be transmitted, wherein the objective function is as follows:
4.3 If s 2≥60-s1), jumping to step 4.5); if the current block number s 2＜60-s1, the Q matrix is updated according to the above objective function, and a white noise matrix WN (1, 60) is randomly generated, multiplied by Adding the third block s 3 with the second row of the Q matrix, and selecting the column number with the maximum value after adding as a third block s 3 of information to be transmitted;
4.4 If s 3≥60-s1-s2), jumping to step 4.5); if the current block s 3＝60-s1-s2 is complete, the Q matrix is updated according to the following objective function:
4.5 Setting the current reward as-1, updating the Q matrix according to the objective function, and repeating the steps 4.1) to 4.4) until the current state is that the information X 3 to be transmitted is blocked, and finishing one-time exploration;
4.6 Repeating the steps 4.1) to 4.5) until the exploration times reach K, and finishing the updating of the Q matrix.
5) And 4) selecting a partitioning scheme according to the updated Q matrix in the step 4). The AUV calculates the analog beam forming matrix according to the optimized sub RIS block, and the calculation is completed through the phase shifter, so that the purpose that the sound wave transmits information to a plurality of destination nodes through the RIS system at the same time is achieved.
6) After the RIS system completes phase regulation, the AUV transmits a preparation signal to the information source node SN, and the SN integrates the information sub-packets to be transmitted and then sends the integrated information sub-packets to the AUV. The information sub-packet is reflected to a plurality of destination nodes through the RIS system, and the destination nodes process the signal segments belonging to the destination nodes according to the signal labels.
The feasibility of the method of the invention is verified by computer simulation.
Simulation using MATLAB R2021b platform. The simulation parameters were set as follows: the underwater acoustic network is set to be a 1000×1000 two-dimensional scene, and comprises s=50 sensor nodes and one information source node, wherein the information source node SN is positioned at a coordinate (500,0); the range H of the energy cavity is set to be a circular range with a radius of 300m around (500 ), namely, an underwater sound network schematic diagram shown in fig. 2. An aus system equipped with 60×60 was set up on the AUV. The SN has x=3 pieces of information to be transmitted, and the information importance of the three pieces of information X 1,X2,X3 is IL i＝1,ILi＝3,ILi =5, which are to be transmitted to destination nodes (250, 1000), (500, 1000), (750, 1000) respectively. The Q-learning maximum number of heuristics is set to k= 7000000.
Fig. 3 is a schematic diagram of rewards under different block behaviors, showing that in the transmission of information with three information importance degrees of X 1,X2,X3 being IL i＝1,ILi＝3,ILi =5, rewards of different block number actions are selected under the Q-learning algorithm, and it can be seen that under the trade-off between the beam forming capability and the computational complexity, the more the block number is, the better the reward function is presented as an upward convex curve which increases and decreases with the block number. As can be seen from fig. 3, the gain of the information block number increase with low information importance is lower than that of the information block number increase with high importance, i.e. the slope of the curve with high importance is larger.
FIG. 4 shows a graph of Q-value distribution when the number of Q-learning searches is 7000000. The final blocking result can be obtained as follows: the number of blocks of the information X 1 with the information importance level 1 is 6, the number of blocks of the information X 2 with the information importance level 3 is 10, and the number of blocks of the information X 3 with the information importance level 5 is 44. According to the calculation complexity formula O (N) =n 6, the calculation complexity after the block is O (N) = 6+106+446＝7.25749, and compared with the calculation complexity O (N) =3×60 6＝1.399711 of three times of RIS system forwarding without the block, the calculation complexity after the block is reduced by two orders of magnitude. And three information can be forwarded simultaneously by utilizing the characteristic that the RIS system can perform phase regulation beam forming, and compared with the information forwarding one by one, a large amount of data forwarding time is saved.
Claims (2)
1. An AUV auxiliary underwater sound network data transmission method based on an intelligent super surface is characterized by comprising the following steps:
1) Determining an area H where an energy hole is located in a water sound communication network and a source node SN needing to transmit data, classifying the importance IL i of the water sound data, specifically classifying the information importance IL i of the water sound data into five stages, namely, IL i = i,2,3,4 and 5, wherein the higher the stage is, the higher the information importance is represented; setting X pieces of information to be transmitted on the SN, respectively transmitting the X pieces of information to be transmitted to X destination nodes, and knowing the importance level of the information to be transmitted; the information source node combines the importance level of each piece of information to be transmitted and the position information of the destination node into a preparation data sub-packet PS;
2) The AUV receives the instruction to drive to the center of the energy hole H, and the information source node sends a preliminary data sub-packet PS to the AUV;
3) An N multiplied by N RIS unit system is arranged on the AUV, the total area S=N multiplied by N of the RIS system, and the AUV controls the RIS system to carry out block division and RIS reflection path design according to the prepared data sub-packet PS; according to the information importance of X pieces of information to be transmitted, each piece of information corresponds to a sub RIS block S i(i＝1,2,…,X),Si＝Nxi×Nyi (i=1, 2, …, X), the higher the information importance is, the larger the sub RIS block area S i is, and the sum of the areas of all sub RIS blocks is the total area of the RIS system, namely The Q-learning algorithm is utilized to carry out block optimization on X sub RIS blocks;
4) The AUV calculates an analog beam forming matrix according to the optimized sub RIS block, and the calculation is completed through the phase shifter, so that the purpose that the sound wave simultaneously transmits information to a plurality of destination nodes through the RIS system is achieved;
5) After the RIS system completes phase regulation, the AUV transmits a preparation signal to an information source node SN, and the SN integrates the information sub-packets to be transmitted and then sends the integrated information sub-packets to the AUV; the information sub-packet is reflected to a plurality of destination nodes through the RIS system, and the destination nodes process the signal segments belonging to the destination nodes according to the signal labels.
2. The AUV-assisted underwater acoustic network data transmission method based on intelligent subsurface as claimed in claim 1, wherein in step 3), the specific method for block optimization of X sub RIS blocks by Q-learning algorithm is as follows: first, the function Q (s, a) is initialized, and the function is rewardedLearning rate alpha, attenuation coefficient gamma and training frequency K; s represents the blocking state of the whole RIS system; a represents actions selected by an Agent, namely specific block actions for the RIS system; q (s, a) functions represent state action values, namely benefits obtained by the Agent in the current state by selecting specific block actions, and guide the action selection of each step; reward function/>The method comprises the steps of representing the benefits obtained by the environmental feedback Agent under different states and actions, namely the benefits of different specific block actions of the RIS system; the learning rate alpha determines the updating speed of the Q matrix; the attenuation coefficient gamma determines the importance degree of decision-making on the gain of long-term decision-making;
Reward function The settings were as follows:
Wherein w 1 represents the bonus weight obtained when the sub RIS block is distributed to the size of S i, w 1 is related to the importance of the information to be transmitted, the higher the importance level of the information is, the stronger the required beamforming capability is, the higher the area of the distributed sub RIS block is, and the larger the bonus weight is; w 1 is calculated as:
w1＝β·ILi
Beta represents the bonus weight based on the importance of the information, w 2 represents the bonus weight subtracted from the sub RIS block due to computational complexity; o (N) represents beamforming computational complexity, related to the number of RIS units allocated;
The training starts to randomly select an initial state s, randomly generates a white noise matrix WN, adds the white noise matrix WN with a current Q matrix to obtain a temporary selection matrix, selects an action a 0 corresponding to the maximum value of the temporary selection matrix in all possible actions of the current state s 0, reaches the state s 1 after selecting the action a 0, and updates the Q matrix by using the following formula:
Q(s，a)←Q(s，a)+α[r+γmaxa′Q(s′，a′)-Q(s，a)]
and repeatedly selecting states and actions and updating the Q matrix until training is finished, and obtaining an optimized sub RIS partitioning scheme.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211471729.8A CN115865220B (en) | 2022-11-23 | 2022-11-23 | AUV auxiliary underwater sound network data transmission method based on intelligent super surface |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211471729.8A CN115865220B (en) | 2022-11-23 | 2022-11-23 | AUV auxiliary underwater sound network data transmission method based on intelligent super surface |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115865220A CN115865220A (en) | 2023-03-28 |
| CN115865220B true CN115865220B (en) | 2024-06-11 |
Family
ID=85665222
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211471729.8A Active CN115865220B (en) | 2022-11-23 | 2022-11-23 | AUV auxiliary underwater sound network data transmission method based on intelligent super surface |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115865220B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102323584A (en) * | 2011-05-19 | 2012-01-18 | 哈尔滨工程大学 | A kind of autonomous anticollision device of AUV/UUV platform under water that is applicable to |
| CN113691391A (en) * | 2021-07-13 | 2021-11-23 | 厦门大学 | Underwater acoustic network medium access control method with variable node number based on Q learning |
| WO2021239067A1 (en) * | 2020-05-29 | 2021-12-02 | 华为技术有限公司 | Cooperative communication method and communication apparatus |
-
2022
- 2022-11-23 CN CN202211471729.8A patent/CN115865220B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102323584A (en) * | 2011-05-19 | 2012-01-18 | 哈尔滨工程大学 | A kind of autonomous anticollision device of AUV/UUV platform under water that is applicable to |
| WO2021239067A1 (en) * | 2020-05-29 | 2021-12-02 | 华为技术有限公司 | Cooperative communication method and communication apparatus |
| CN113691391A (en) * | 2021-07-13 | 2021-11-23 | 厦门大学 | Underwater acoustic network medium access control method with variable node number based on Q learning |
Non-Patent Citations (1)
| Title |
|---|
| 陈友淦 ; 许肖梅.人工智能技术在水声通信中的研究进展.《哈尔滨工程大学学报》.2020,全文. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115865220A (en) | 2023-03-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101313494A (en) | Radio network design apparatus and method | |
| CN105635962A (en) | Automated wlan radio map construction method and system | |
| CN114422056A (en) | Air-ground non-orthogonal multiple access uplink transmission method based on intelligent reflecting surface | |
| CN113316169B (en) | UAV auxiliary communication energy efficiency optimization method and device for smart port | |
| CN112511250A (en) | DRL-based multi-unmanned aerial vehicle air base station dynamic deployment method and system | |
| CN114422363A (en) | Unmanned aerial vehicle loaded RIS auxiliary communication system capacity optimization method and device | |
| CN112784362A (en) | Hybrid optimization method and system for unmanned aerial vehicle-assisted edge calculation | |
| CN115134242B (en) | Vehicle-mounted computing task unloading method based on deep reinforcement learning strategy | |
| CN109946673B (en) | Networked radar node selection method based on Monte Carlo tree search | |
| CN114885340B (en) | Ultra-dense wireless network power distribution method based on deep migration learning | |
| Gong et al. | When optimization meets machine learning: The case of IRS-assisted wireless networks | |
| Gupta et al. | Multi-UAV deployment for NOMA-enabled wireless networks based on IMOGWO algorithm | |
| Yang et al. | Secure and cooperative target tracking via AUV swarm: A reinforcement learning approach | |
| CN115865220B (en) | AUV auxiliary underwater sound network data transmission method based on intelligent super surface | |
| CN112738849B (en) | Load balancing regulation and control method applied to multi-hop environment backscatter wireless network | |
| CN113676979A (en) | End-to-end routing relay link selection method, system and device based on branch-and-bound method | |
| Surajudeen-Bakinde et al. | Effect of membership functions and data size on the performance of ANFIS-based model for predicting path losses in the VHF and UHF bands | |
| CN117424633A (en) | Secure communication transmission strategy and system under deep learning auxiliary active ARIS | |
| CN117295090A (en) | Resource allocation method for Unmanned Aerial Vehicle (UAV) through-sense integrated system | |
| CN112329371A (en) | Inverse modeling method of Doherty power amplifier based on IALO-BP neural network | |
| CN116405493A (en) | Edge cloud collaborative task unloading method based on MOGWO strategy | |
| Gao et al. | Multi-objective optimization of energy and latency in URLLC-enabled wireless VR networks | |
| KR102555696B1 (en) | Device and method for allocating resource in vehicle to everything communication based on non-orhthogonal multiple access | |
| Wang et al. | Active-IRS-Enabled Energy-Efficiency Optimizations for UAV-Based 6G Mobile Wireless Networks | |
| Xu et al. | Constrained multi-objective task assignment for UUVs using multiple ant colonies system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |